Image stitching on the unmanned air vehicle in the indoor environment

This paper presents an image stitching algorithm for applied on the unmanned air vehicle in the indoor environment. Since the images are captured on this moving platform, the sudden vibration may cause the images blurred. These unclear frames must be removed for improving the quality of the mosaic image. Also, when stitching the images in the indoor environment, the distance between the scene and camera is closer than that in the outdoor air reconnaissance mission. The motion parallax easily exists in the continuously captured image sequence. The hierarchical image stitching method with the dominant image selection for a same scene is proposed to reduce the motion parallax region in the mosaic image. The experimental results validate the proposed image stitching algorithm producing a mosaic image with better image quality than that of the other method.